AI Engineer (LLM & Agents)

ECPK IT

  • Алматы
  • Постоянная работа
  • Полная занятость
  • 7 ч. назад
We're looking for a builder who's equally comfortable prototyping a new agent architecture and deploying it to production. If you've been tinkering with LLMs, building RAG pipelines, and dreaming about what AI can do for fintech—this is your chance to make it real.Responsibilities:
  • Design, build, and maintain LLM-based applications and agentic systems that solve real business problems across the organization;
  • Develop and optimize RAG pipelines for internal knowledge retrieval, document processing, and multilingual content;
  • Build and deploy AI agents capable of autonomous task execution, tool use, and multi-step reasoning;
  • Integrate LLM capabilities with existing systems via APIs and MCP (Model Context Protocol), working with payment platform data and business workflows;
  • Evaluate, fine-tune, and benchmark models for specific use cases (cost, latency, accuracy trade-offs);
  • Maintain and extend our AI platform infrastructure (Open-source web based user interfaces, AWS Bedrock, embedding models, vector databases);
  • Collaborate with business stakeholders to identify high-impact AI opportunities and translate them into working solutions;
  • Stay current with the rapidly evolving LLM/agent landscape and bring relevant innovations into our stack.
Requirements:
  • 3+ years of software engineering experience, with at least 1 year focused on LLM applications
  • Hands-on experience building with LLM APIs (OpenAI, Anthropic Claude, AWS Bedrock, or similar);
  • Practical experience implementing RAG systems: embeddings, vector stores, retrieval strategies, chunking approaches;
  • Familiarity with agent frameworks and patterns (function calling, tool use, ReAct, multi-agent systems);
  • Strong Python skills; comfortable with async programming, API development, and data pipelines;
  • Experience with cloud infrastructure (AWS preferred: Lambda, ECS, Bedrock, S3);
  • Understanding of prompt engineering principles and LLM evaluation methodologies;
  • Self-directed problem solver who thrives with autonomy and minimal hand-holding;
  • Fluent English (written and spoken);
Nice to have:
  • Experience with LangChain, LlamaIndex, or similar orchestration tools;
  • Background in payments, fintech, or e-commerce domains;
  • Experience with Kubernetes and containerized deployments;
  • Familiarity with fine-tuning or training models (LoRA, PEFT, etc.);
  • Knowledge of compliance/security considerations for AI in regulated industries.
Why It’s Exciting to Work With Us:
  • Whether you want to grow as a leader, deepen your expertise, or explore new roles, we're here to support you every step of the way;
  • You choose the tools and techniques that will help you achieve your goals;
  • Ownership and freedom to drive your ideas and decisions.
ECOMMPAY – международный платежный провайдер и прямой эквайер. Мы обеспечиваем проведение онлайн-платежей и выплат для среднего и крупного онлайн-бизнеса по всему миру.ECOMMPAY предлагает индивидуальные платежные решения с учетом потребностей каждого клиента. Наши технологии основаны на детальном анализе сферы деятельности предприятия, индустрии, рынка и разработаны с учетом всех рисков.ECOMMPAY предоставляет глобальное покрытие с локальной экспертизой. Более 100 методов оплаты, отказоустойчивая IT инфраструктура с показателем 99,99% и собственная система управления рисками с уровнем выявления фрода – 97%. Более того, ECOMMPAY – один из первых европейских провайдеров, начавших подключать своих клиентов к инновационной платежной технологии Open Banking. Компания была основана в Лондоне в 2012 году. Сейчас у ECOMMPAY 4 офиса по всему миру, в которых работают более 700 сотрудников, включая более 350 ИТ-специалистов.

HeadHunter